How should 343 Make Fair Multiplayer Matches Work?

So i don’t know about anyone else but my previous experiences with multiplayer gameplay in any game, is every 3rd or so game i play i find people who are MUCH better then me to the point where it doesn’t even seem fair, How should 343i deal with this problem? To make the multiplayer experience fun and enjoyable for all. Basically what i’m saying is how should 343i make it so we vs other people with a similar skill level.

My idea go off K/D spread and if your in a party with someone who has a higher K/D spread then you, you’ll find people within the difference of ur K/D so you won’t find anyone who’s extremely higher then your friend and no one will be too much lower then you.

The last few games have all tried to match you based on your skill. In Halo 3, it was visible to you but in the games since it was hidden under the hood to try and stop exploits.

The problem is that the matchmaking system has to try and find you the best game possible which requires ideal skill matches and the best connections along with other factors. There’s also a matter of population; there isn’t always enough people to get you into a good game, so it loosens the parameters to try and find the next best thing.

They’ve already got a decent ranking system in place check it out on google just type in halo 5 ranking system

> 2533274839968636;3:
> They’ve already got a decent ranking system in place check it out on google just type in halo 5 ranking system

I already tried it in the beta, i still found people 1-2 full ranks higher then me.

> 2533274923639240;4:
> > 2533274839968636;3:
> > They’ve already got a decent ranking system in place check it out on google just type in halo 5 ranking system
>
>
> I already tried it in the beta, i still found people 1-2 full ranks higher then me.

hmm does seem like a problem hope its fixed by launch or atleast in a patch

KD ratio. That’s what truly determines skill. With the concept of Win/Lose determining skill, it can be skewered. For example, somebody that sucks could have a competitive rank of 20 because they get teammates that always win, or it is someone really good that is stuck with people who always lose, therefore bringing down his/her competitive rank.

> 2533274923639240;1:
> So i don’t know about anyone else but my previous experiences with multiplayer gameplay in any game, is every 3rd or so game i play i find people who are MUCH better then me to the point where it doesn’t even seem fair, How should 343i deal with this problem? To make the multiplayer experience fun and enjoyable for all. Basically what i’m saying is how should 343i make it so we vs other people with a similar skill level.
>
> My idea go off K/D spread and if your in a party with someone who has a higher K/D spread then you, you’ll find people within the difference of ur K/D so you won’t find anyone who’s extremely higher then your friend and no one will be too much lower then you.

No, first thing that tips that idea over is that good players who face off most likely will have K/D spreads similar to that of worse players facing off.

Johnny No-Aim and Amanda Strafe-less are equally good, they face off and should end up with similar K/Ds, close to 1.
Nick Nader and Lisa Trapper are also equally good, but infinitely better than Johnny and Amanda. Nick and Lisa play against each other and also end up with K/Ds close to 1.

See what’d happen when you start putting players against each other based on K/D?

The players you currently see who have a high K/D would need to play against worse players in order to get a high K/D. When they then play against others with a high K/D, it’s lowered and they’re back to playing lesser skilled players again. You’d play them when they get high K/Ds, and you’d play them when they’re ultimately poked down again.

> 2533274859413227;6:
> KD ratio. That’s what truly determines skill. With the concept of Win/Lose determining skill, it can be skewered. For example, somebody that sucks could have a competitive rank of 20 because they get teammates that always win, or it is someone really good that is stuck with people who always lose, therefore bringing down his/her competitive rank.

As I explained above, using K/D to match players, would mean that your K/D should normalize around 1.

I’m sure someone will be around to explain why W/L is superior to tracking and using individual stats.

> 2533274795123910;7:
> > 2533274923639240;1:
> > So i don’t know about anyone else but my previous experiences with multiplayer gameplay in any game, is every 3rd or so game i play i find people who are MUCH better then me to the point where it doesn’t even seem fair, How should 343i deal with this problem? To make the multiplayer experience fun and enjoyable for all. Basically what i’m saying is how should 343i make it so we vs other people with a similar skill level.
> >
> > My idea go off K/D spread and if your in a party with someone who has a higher K/D spread then you, you’ll find people within the difference of ur K/D so you won’t find anyone who’s extremely higher then your friend and no one will be too much lower then you.
>
>
> No, first thing that tips that idea over is that good players who face off most likely will have K/D spreads similar to that of worse players facing off.
>
> Johnny No-Aim and Amanda Strafe-less are equally good, they face off and should end up with similar K/Ds, close to 1.
> Nick Nader and Lisa Trapper are also equally good, but infinitely better than Johnny and Amanda. Nick and Lisa play against each other and also end up with K/Ds close to 1.
>
> See what’d happen when you start putting players against each other based on K/D?
>
> The players you currently see who have a high K/D would need to play against worse players in order to get a high K/D. When they then play against others with a high K/D, it’s lowered and they’re back to playing lesser skilled players again. You’d play them when they get high K/Ds, and you’d play them when they’re ultimately poked down again.
>
>
>
>
> > 2533274859413227;6:
> > KD ratio. That’s what truly determines skill. With the concept of Win/Lose determining skill, it can be skewered. For example, somebody that sucks could have a competitive rank of 20 because they get teammates that always win, or it is someone really good that is stuck with people who always lose, therefore bringing down his/her competitive rank.
>
>
> As I explained above, using K/D to match players, would mean that your K/D should normalize around 1.
>
> I’m sure someone will be around to explain why W/L is superior to tracking and using individual stats.

Ok i see where ur going with this, but whats ur idea of fixing this problem then how can we have people fairly matched?, cause too me and a lot of others K/D seems like the best choice of making everything fair.

> 2533274923639240;8:
> > 2533274795123910;7:
> > > 2533274923639240;1:
> > >
>
>
> Ok i see where ur going with this, but whats ur idea of fixing this problem then how can we have people fairly matched?, cause too me and a lot of others K/D seems like the best choice of making everything fair.

W/L ranking and stricter search criteria.
A good healthy population helps as well.

You and those who see K/D as a good or even the best choice to make everything fair want that arbitrary number to have some sort of meaning, especially in terms of skill, when it sort of has little meaning. What you’re missing is that if you were to match players according to their K/D then good players would still be matched with players worse than them, because their K/D will be lower when they play against each other than when they play against you. You do not get a high K/D from playing against players in your own actual skill range. You get a high K/D by playing against players worse than you.

Let’s take a scenario that might be better understood.

Players and skill

Player 1: 5
Player 2: 5
Player 3: 10
Player 4: 10
Player 5: 20
Player 6: 20

First match is a random set up, because no one has been assigned any sort of value to their skill by the game, because there’s no data to use and the system can’t evaluate a player without data.

Players 5 and 6 are matched against each other: Result is that both have a K/D close to 1
Players 1 and 4 are matched. Result is that Player 1 nets a K/D of 0.5 while player 4 gets 1.5
Players 2 and 3 then get the same result as 1 and 4.

So after that it’s like this:

Player 1: 0,5
Player 2: 0,5
Player 3: 1,5
Player 4: 1,5
Player 5: 1
Player 6: 1

According to that, 3 and 4 are better than 5 and 6, despite the latter two in reality being better. Then, let’s look at the next matches

1 and 2: Both nets a K/D close to 1 as they’re close to each other in skill
3 and 4, again, both nets close to 1
5 and 6 does the same.

If we take the average,

Player 1: 0,75
Player 2: 0,75
Player 3: 1,25
Player 4: 1,25
Player 5: 1
Player 6: 1

See, they’re getting closer. Once they’re in range of each other, they’ll start getting matched against each other and you’ll have unfair matches again, like the first match. You can try different intitial match ups and then a few iterations where the ones with the closest K/Ds are matched, with results being close to each other if the players are equally skilled.

There’s however more to it than just it being a bad way to measure skill, in a team environment, it encourages bad behaviour, team killing if you will, selfish playstyles. Why? Because you benefit from being selfish. You are in a game where there are some other players you shouldn’t shoot at, but you benefit from it if they have a power weapon you can use.

Halo: Reach Arena used individual stats to rank players, something Bungie replaced with a pure W/L system because the Individual system wasn’t good, for those above reasons.

Problems with uneven matches?
Stricter search criteria.
Better population, something i343 can’t do much about though.

Think it’s unfair that a bad player can get a high rank by being carried?
Do what Starcraft 2 does, the team/group gets a rank instead of the individual player.

Also, that player is doing something right as well, your rank is proof of your success at winning. If you play to win you’ll do everything you’re allowed to do, getting a good team is something you’re allowed to do.

Your idea makes no sense lol.

If I play 10 games and have a 3.00 k/d and start getting matched with other kids with 3.00 k/ds after another 10 games my K/D will probably be around 1.00. Then i’ll play kids with 1.00 K/Ds and after ten games i’ll be back to 3.00 K/D lol.

You can’t match people based on k/d because k/d is relative to who you are playing.

You drink a kick start ( hope fully a double xp for halo 5)

Make sure to blame lag , them having faster internet and now you can blame that you weren’t on red team

> 2533274839968636;5:
> > 2533274923639240;4:
> > > 2533274839968636;3:
> > > They’ve already got a decent ranking system in place check it out on google just type in halo 5 ranking system
> >
> >
> > I already tried it in the beta, i still found people 1-2 full ranks higher then me.
>
>
> hmm does seem like a problem hope its fixed by launch or atleast in a patch

Probably wouldn’t be totally balanced in just under two weeks for how long the Beta lasts. It takes time to rank and sort players accordingly.

I think the system would work perfectly if it would consider both similar KD ratio and WL ratio!

> 2533274923639240;4:
> > 2533274839968636;3:
> > They’ve already got a decent ranking system in place check it out on google just type in halo 5 ranking system
>
>
> I already tried it in the beta, i still found people 1-2 full ranks higher then me.

I didn’t like the H5 rank system, it seemed like it was more based on winning games than it was personal skill, which is dumb since most average around 50% wins/losses. I noticed it just kept swinging back and forth until I went on a streak one night and just kept winning game after game. I finally ranked up - once.

Finding a game that puts you with people of the same exact rank is a pipedream, and not realistic. It would take forever to get into a match.

How well a search places you with like ranks is all based on playlist population. If it’s high, you’re likely to get into a match with similiar ranks. If it’s low, the search has to broaden it’s search to a greater variation for the sake of time. It is what it is.

I always thought k/d ratio should play a significant role in ranking as well, but that would have it’s own problems, mainly the fact that your k/d for one match has a lot to do with the skill of the other players, players that are often better or worse. Put everyone in games with like skilled players and everyone’s k/d is going to hover around 1 now. Now k/d is pointless.

Halos 2 and 3 matched you up with people who were at your skill level. 5 will do the same.

> 2533274859413227;6:
> KD ratio. That’s what truly determines skill. With the concept of Win/Lose determining skill, it can be skewered. For example, somebody that sucks could have a competitive rank of 20 because they get teammates that always win, or it is someone really good that is stuck with people who always lose, therefore bringing down his/her competitive rank.

I disagree entirely. Most objective modes are going to bring your K/D down for obvious reasons. In Slayer, what if you’re just getting unlucky and getting assists instead of kills for a match or two? What if I’m still helping my team by assisting them with kills and then I rank down just because I didn’t go positive? That’s not fair, I still helped my team out even if I didn’t go positive. As far as I’m concerned, Objective modes should only look at W/L and Slayer modes should take in W/L as well as your individual stats.

If I constantly win and never lose, I should never rank down, even if I go negative during a match. But if someone on my team does better than me in the match, he should rank up faster than I do.

> 2533274795123910;7:
> > 2533274923639240;1:
> > So i don’t know about anyone else but my previous experiences with multiplayer gameplay in any game, is every 3rd or so game i play i find people who are MUCH better then me to the point where it doesn’t even seem fair, How should 343i deal with this problem? To make the multiplayer experience fun and enjoyable for all. Basically what i’m saying is how should 343i make it so we vs other people with a similar skill level.
> >
> > My idea go off K/D spread and if your in a party with someone who has a higher K/D spread then you, you’ll find people within the difference of ur K/D so you won’t find anyone who’s extremely higher then your friend and no one will be too much lower then you.
>
>
> No, first thing that tips that idea over is that good players who face off most likely will have K/D spreads similar to that of worse players facing off.
>
> Johnny No-Aim and Amanda Strafe-less are equally good, they face off and should end up with similar K/Ds, close to 1.
> Nick Nader and Lisa Trapper are also equally good, but infinitely better than Johnny and Amanda. Nick and Lisa play against each other and also end up with K/Ds close to 1.
>
> See what’d happen when you start putting players against each other based on K/D?
>
> The players you currently see who have a high K/D would need to play against worse players in order to get a high K/D. When they then play against others with a high K/D, it’s lowered and they’re back to playing lesser skilled players again. You’d play them when they get high K/Ds, and you’d play them when they’re ultimately poked down again.
>
>
>
>
> > 2533274859413227;6:
> > KD ratio. That’s what truly determines skill. With the concept of Win/Lose determining skill, it can be skewered. For example, somebody that sucks could have a competitive rank of 20 because they get teammates that always win, or it is someone really good that is stuck with people who always lose, therefore bringing down his/her competitive rank.
>
>
> As I explained above, using K/D to match players, would mean that your K/D should normalize around 1.
>
> I’m sure someone will be around to explain why W/L is superior to tracking and using individual stats.

Then we have Cory No Objective and Joseph Spawn Trapper, with over inflated KD ratios, and do nothing to help the team win.

> 2533274833600810;17:
> > 2533274859413227;6:
> > KD ratio. That’s what truly determines skill. With the concept of Win/Lose determining skill, it can be skewered. For example, somebody that sucks could have a competitive rank of 20 because they get teammates that always win, or it is someone really good that is stuck with people who always lose, therefore bringing down his/her competitive rank.
>
>
> I disagree entirely. Most objective modes are going to bring your K/D down for obvious reasons. In Slayer, what if you’re just getting unlucky and getting assists instead of kills for a match or two? What if I’m still helping my team by assisting them with kills and then I rank down just because I didn’t go positive? That’s not fair, I still helped my team out even if I didn’t go positive. As far as I’m concerned, Objective modes should only look at W/L and Slayer modes should take in W/L as well as your individual stats.
>
> If I constantly win and never lose, I should never rank down, even if I go negative during a match. But if someone on my team does better than me in the match, he should rank up faster than I do.

Agree, this right here is best way. Everything you just said is what I was about to type. There are players that are also good at call outs and stuff like that but they might not be the most weapon skilled player so they don’t get crazy amount of kills . However those players are still play a big role on winning the game.